Automated critical power limit estimation for natural convection-cooled research reactor core
Introduction
Currently, more than two hundred research reactors are in operation worldwide, with varying core powers owing to their diverse utilization purposes*** (IAEA, 2019). The core heat is removed by forced cooling by pump or by natural convection. For a TRIGA (Training, Research, Isotopic Production General Atomic) reactor, it is reported that up to 2 MW of core thermal power can be safely removed by natural convection (IAEA, 2018). Based on this criterion, about 69% of the research reactors worldwide have cores that may be passively cooled. Some of the well-known research reactors with natural convective core cooling are TRIGA, SLOWPOKE, and MNSR. TRIGA was developed by General Atomics, and there are three design variations (Mark-I to Mark-III). For different utilization requirements, its design core power can be varied from several tens of kW to several MW. Most of the core utilizes rod-type U-ZrH fuel with top and bottom graphite reflectors (Stancar and Snoj, 2017). SLOWPOKE (Safe LOW-POwer Kritical Experiment) was designed by Atomic Energy of Canada Limited, where prototype (SLOWPOKE-1) and commercialized (SLOWPOKE-2) versions exists. The core thermal power of SLOWPOKE-2 is 20 kW, and rod-type fuel made of UO2 meat and Zircaloy-4 cladding is currently used (Townes and Hilborn, 1985). The MNSR (Miniature Neutron Source Reactor) was developed by China Institute of Atomic Energy. It has similar design concepts as SLOWPOKE-2. To remove 30 kW of core heat, the coolant flows into the core through the gap between the lower reflector and annular beryllium block. Its rod-type fuel utilizes U-Al alloy meat (Yongmao, 1986, Dawahra, 2014). Korea Atomic Energy Research Institute (KAERI) is also developing a low-power research reactor (LPRR) for training and educational purpose, where rod-type and low enriched uranium oxide fuel is utilized to generate 50 kW of thermal power (Kim, 2018).
Unlike forced cooled research reactors, the heat of the natural convection-cooled core is removed by buoyancy-induced upward flow, which means the flow rate is coupled with the power and not directly controllable. Since the core equilibrium flow rate is not linearly proportional (mass flow ∝ power1/3) to the thermal power, the amount of power increase cannot be fully compensated by the flow increase (Lewis, 1977). In the case of an operator malfunction or a reactivity-induced accident, the core power may increase. The power level then may reach the critical value (critical power) at which the local wall heat flux corresponds to the critical heat flux (CHF) and fuel failure can occur. The reactor should be tripped by any means before reaching this limit to bring the reactor to safety. This critical core power limit is used as a base value to establish safety limits by including conservatism and transient effects (IAEA, 2008, McGuire, 1995). Because finding this kind of limit is quite a rigorous task, this study developed an automated method to generate a power limit map for the natural convection-cooled research reactor core. First, the variation range of the major thermal-hydraulic parameters was found from steady-state and transient analysis results. A literature review was conducted to identify applicable CHF correlations and their design limits for estimating critical power. An in-house companion code RCPP (RELAP5 Companion Processor) was developed to control the batch run of the system code RELAP5/MOD3.3 and to post-process the results. The RELAP5/MOD3.3 is the system thermal hydraulic analysis code, which is widely used in the nuclear industry (USNRC, 2001). For each run, this code calculated thermal hydraulic variable values of the reactor core region necessary for the CHF ratio evaluation. Using the RELAP5/MOD3.3 and the RCPP, the critical power limit was found for each combination of the major operation parameters, namely, the core inlet temperature and pool height. Finally, a critical power map was created by merging the critical power limits for the range of operating conditions.
Section snippets
Problem description
In this study, a fuel assembly of the LPRR being developed by KAERI was modeled and analyzed to construct a critical core power limit map (Kim, 2018).
Correlation applicability
Table 3 summarizes candidate correlations after preliminary screening. The correlations were screened based on four selection criteria (general correlation, reputation, embedded in system analysis code, and has error statistics). An AECL lookup table (LUT) is a collection of CHF values for local thermal hydraulic conditions (pressure, mass flux, and quality) normalized for 8 mm tube geometry. Several correction factors were applied to consider geometric and flow-related effects (Groeneveld et
Computational fluid dynamic simulation
As seen in Fig. 1, the heat from the fuel rods is removed by coolant that flow through holes in the grid plate and core box wall. In this study, only the flow through grid plate holes were considered in the analysis, because reducing the number of possible inflow paths simplifies the problem and yields more conservative (higher) temperature distribution. Before preceding to the system analysis, a computational fluid dynamic (CFD) simulation was carried out using commercial code CFX 16.1 to
Critical core power limit map
The critical core power limit map was generated by combining predicted critical values for a wide range of thermal hydraulic boundary conditions (inlet temperature = 25 ~ 95 °C, pool height = 6 ~ 8 m) which covers the normal operation range. As shown in Table 7, maps were created for each combination of the CHF correlation and the core state. Fig. 19 shows contour maps of the resulting critical powers. Overall, as the pool water height is increased and the core inlet temperature are decreased,
Conclusions
In this study, a critical power limit map generation method was developed for the natural circulation-cooled research reactor core. In brief, this method is divided into two stages. The first stage deals with the constructing system code input for MCHFR calculation, which requires the correlation limit to be found from suitable CHF correlation and EHCF values. The next stage is about finding critical power values for given boundary conditions and convergence criteria. To automate this second
CRediT authorship contribution statement
Hyung Min Son: Conceptualization, Methodology, Software, Formal analysis, Writing - original draft. Jonghark Park: Validation, Formal analysis, Investigation, Visualization, Supervision, Project administration.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
This work was conducted as a part of the Development of Research Reactor Technology project sponsored by the Ministry of Science and ICT of the Korean government.
References (32)
An Independent Assessment of Groeneveld et al’.s 1995 CHF Look-up Table
Nucl. Eng. Des.
(1997)Determination of the LEU core safety parameters of the MNSR reactor using the MCNP4C code
Ann. Nucl. Energy
(2014)On the definition of critical heat flux margin
Nucl. Eng. Des.
(1996)The 1995 look-up table for critical heat flux in tubes
Nucl. Eng. Des.
(1996)The 2006 CHF look-up table
Nucl. Eng. Des.
(2007)- et al.
Critical Heat Flux (CHF) for water flow in tubes-II. Subcooled CHF correlations
Int. J. Heat Mass Transfer
(2000) - et al.
An improved thermal power calibration method at the TRIGA Mark II research reactor
Nucl. Eng. Des.
(2017) Heat transfer from tubes in crossflow
Adv. Heat Transfer
(1972)- “ANSYS CFX Introduction, Release 16.1,” ANSYS Inc., Canonsburg, PA,...
- “ANSYS CFX-Solver Theory Guide, Release 16.1,” ANSYS Inc., Canonsburg, PA,...